Paper1

1

CONCRETE-COMPOSITE BEAM-COLUMN JOINTS-

2

I. PERFORMANCE OF FRESH SPECIMENS

3

A. Mukherjeea *, G. L. Raib a

4

Director, Thapar University

5

Patiala 147004, India

6

Tel: +91 175 2393001, 2363007 Fax: +91 175 2364498, 2393005

7

b

Research scholar, Dept. of Civil Engineering, Indian Institute of Technology Bombay,

8

Mumbai 400076, India

9

Tel: +91 9322597149, 22 32610117

10

E-mail: a [email protected], b [email protected]

11 12

ABSTRACT

13The present paper discusses the performance of reinforced concrete (RC) beam-column joints 14under cyclic excitation. Two types of beam-column joints, with ductile and brittle reinforcement 15detailing, have been cast. The joints are subjected to cyclic displacements of monotonically 16increasing amplitude until failure. Post failure, the joints have been rehabilitated using reinforced 17polymer composites (FRPC). The rehabbed joints have been subjected to the same load regime. 18The performance of rehabbed joints has been compared with that of the fresh joints. The 19investigation highlights the efficiency of the proposed rehabilitation scheme in enhancement of 20strength and deformability of joints. In this part of the paper the performance of the fresh joints 21has been discussed. 22 23Keywords: Reinforced concrete, beam-column Joints, cyclic-loading, ductility, failure 24patterns, energy dissipation. 1 2

1

1 2

INTRODUCTION

3Although hundreds of thousands of successful RC framed structures are annually constructed 4worldwide, there are large numbers of them that deteriorate, or become unsafe due to changes in 5loading, changes in use, or changes in configuration. Occurrence of natural calamities may also 6render a large number of structures unusable. Failure of structures at the beam-column joints in 7the recent earthquakes has exposed the lacuna in the building codes in recommending adequate 8reinforcements and their proper detailing at the joints. The lack of confinement of concrete due 9to sparse spacing of links has caused crushing of concrete (Fig. 1a). The lack of adequate links 10has also led to shear failure at the joints (Fig. 1b). The inadequate bond length of reinforcements 11and improper lapping has been the cause of local weakness and failure (Fig. 1c). 12The pitfalls of improper reinforcement detailing have been highlighted by many researchers. In 13the following discussion only key observations relevant to the present objectives have been 14included. The provision of adequate spirals or hoop reinforcements at critical sections was 15suggested for resisting the bursting pressure due to compression in concrete as well as the 16tension from the beam reinforcements [1-3] Importance of bond between the longitudinal bars 17and concrete for dissipation of energy has been highlighted.[4-5] The combined effect of joint 18shear and the compressive load on the column has been studied and it was observed that higher 19column compression delays shear cracking [6-7] Based on the experiments of a wide range of 20scales of joints it is observed that the small scale samples had faster deterioration of stiffness due 21to early loss of bond [8] 22Although the importance of joint reinforcements to improve the joint ductility has been 23emphasized in research and design codes there are a large number of structures where the ductile 24detailing has not been followed. Therefore, it is important to develop strategies of retrofitting and 1 2

2

1rehabilitation of such joints. Retrofitting deficient joints that have not been damaged has been 2investigated [9-11]. Recently, rehabilitation technique for complete recovery of joints that have 3damaged to the extent that they have exhausted their moment resistance has been reported [12]. 4The work was restricted to small scale joints. The objective of this paper is to develop a 5rehabilitation strategy of damaged beam-column joints of representative size. The performance 6of RC beam-column joints, both ductile and brittle, has been reported. Fresh specimens have 7been subjected to cyclic deformation of monotonically increasing amplitude until the load 8resistance of the joint is completely exhausted. The joints have then been rehabbed with the 9proposed technique. The rehabbed specimens have been subjected to the same load regime. The 10part I of the paper reports the performance of the fresh joints. The performance of the proposed 11rehabilitation technique is discussed in the second part of the paper. 12 13

EXPERIMENTAL PROGRAM

14The present experiment has four stages- specimen preparation, infliction of damage, 15rehabilitation and final tests. In the first part of the paper the first two phases have been 16described. 17 Material System 18The material system consists of concrete, steel reinforcements and CFRP sheets and strips. The 19concrete mix was designed for target strength of 30MPa on 150-mm cubes. The cubes have been 20cast from each batch of concrete and a record of their strength after 28 days of curing has been 21kept. The average strength of the cubes was 32MPa the standard deviation was ± 1.2 and stress22strain curve is shown in Fig.2. 23

1 2

3

1The reinforcement steel for both longitudinal and transverse reinforcement was tested for tensile 2capacity. The longitudinal steel had average yield strength of 650MPa and the steel for stirrups 3had yielded at an average stress of 500MPa. Table 1 presents the details of the reinforcements. 4 5Test Specimens 6The configuration and dimensions of the joints are shown in Fig. 3. All the specimens had 7identical dimensions. They had the same longitudinal reinforcements in the beams and columns. 8However, the spacing and the position of stirrups differed in the two sets. One set had closely 9spaced stirrups (75mm) to provide adequate confinement and shear capacity. The stirrups were 10provided in the core joint region as per the contemporary ductile detailing practice (D-type). In 11the other set of specimens the spacing of stirrups is relatively sparse (150mm) as per the past 12practice that would lead to brittle failure (B-type). In this case, there is no stirrup in the core joint 13region. There is a large stock of facilities built according to the old codes and the objective of 14this test is to examine the proposed rehabilitation technique for those structures. While the first 15group of specimens should not fail in shear the second group is deficient in shear capacity. 16 17The reinforcement cages were prepared taking care of the precise position of the longitudinal 18bars and stirrups. They were placed in steel molds and a cover of 20mm was provided. The 19position of the reinforcement cage was maintained by means of spacers. The standard M30 grade 20concrete mix was used. The specimens are demolded after 24 hours and kept in the curing tank 21for 28 days. The surface dried specimens were used in testing. 22 23 Damage Infliction

1 2

4

1The experimental setup is shown in figure 4a. The specimen was fixed at the ends of columns 2and the columns were subjected to a constant axial load by means of hydraulic jacks. The 3magnitude of load was monitored through a load cell (Fig. 4a). To observe the effect of column 4axial load on the behavior of the joint the samples had different column axial forces. The 5columns of the ductile specimens were loaded with 10% of their axial capacity. It has been 6reported earlier that the behavior of shear deficient specimens alters appreciably with the 7variation of axial force in the column. Therefore, the brittle specimens were tested with 10%, 825% and 37.5% column axial force (Table 2). 9To inflict damage the joints were subjected through a predetermined displacement regime. The 10displacement is applied by attaching a dynamic load actuator of 500kN capacity. The initial 11displacement amplitude was 2mm and it was incremented with a step of 2mm in each epoch until 1240mm displacement was reached(Fig. 5). Three identical displacement cycles consisted of one 13epoch of displacement. The accuracy of the displacement measurement was ±0.01mm. The 14frequencny of load was maintained at 1 cycle in four minutes. 15Several displacements and strains have been recorded during the experiment (Fig. 4b). The 16displacements have been recorded by means of linear variable displacement transducers (LVDT) 17of travel ±150mm. LVDT1 measures the tip dispalcement of the loaded beam. LVDT2 and 18LVDT4 measure the horizontal deformation of the centres of the columns. From these readings 19we can estimate the rotation of the column at the joint. LVDT3 measures the horizontal 20displacement of the center of the joint. If all the loads are applied vertically and the ends are 21restrained properly there should not be any displacement of the center in the horizontal direction. 22There was no appreciable displacement recorded by LVDT3. LVDT5 was placed perpendicular 23to the plane of the joint. This was done to ensure absence of any out-of-plane displacement of the 24specimen at the time of loading. The strains have been measured at the critical locations on steel 1 2

5

1bars and CFRP strips. The responses have been recorded online using a data acquisition system. 2However, the strain gages were affected at the onset of damage in the specimen. Therefore, only 3limited results could be recorded by the strain gages. The cracks on the surface of the specimens 4were marked at the completion of each load cycle and they were traced on a grid paper. Widths 5and paths of the cracks were noted after each cycle. 6It may be noted that the specimens had two arms. The arms were loaded sequentially; i.e. after 7one arm is subjected to the entire load regime the other arm was loaded. The test on the second 8arm yielded the data on the performance of partially damaged joints. 9

1 2

6

1

EXPERIMENTAL RESULTS

2The parameters that have been monitored in the present investigation are the load-deflection 3hysteresis of the tip of the beam, the history of cracks, and the failure modes. 4 5Ductile vs Brittle Joints 6The hysteresis of Arm1 of the ductile and the brittle joints for 10% column compression is 7presented in Fig. 6. It is clear from the hysteretic loops that although the peak load in both the 8specimens was comparable, D specimens show gradual decrease in stiffness, while B specimens 9show sudden decrease in stiffness after reaching the peak value. To compare the performance of 10the two specimen types the envelopes of the two load-displacement graphs are plotted in Fig. 7. 11The ductile and the brittle specimens behave quite similarly until the yield. Postyield, while the 12brittle specimen degrades rapidly the ductile specimen retains a large proportion of its strength 13leading to a graceful failure. 14 15To understand the improved postyield behavior the history of appearance of cracks in the two 16specimens is studied (Fig. 8). The position of reinforcements is marked in dotted lines. The first 17link in the beam in the D specimens was at 50mm from the column longitudinal reinforcement 18and at 75mm spacing thereafter. In B specimens the first link was at the face of the column and 19the spacing was 150mm. Another notable difference is that in the D joint there is one column 20link at the center of the joint while it is absent in the B joint. These details are according to the 21ductile [13] and non-ductile [14] joints of Indian Standards. In Fig. 8 the locations of the links 22serve as reference lines for mapping the cracks. The differently colored solid lines indicate the 23paths of the cracks and the associated number indicates the level of tip displacement at which the 24crack appeared. It may be noted that the first crack appeared at the face of the column at 2mm tip 1 2

7

1displacement in both samples. This is the tension crack in concrete. Therefore, at this 2displacement level the concrete under tension transferred the entire force on to the tensile 3reinforcement. Thereafter, the main difference in the two specimens was the number and 4inclination of the cracks. The cracks were mainly confined in the compartments formed by the 5successive links. In the D samples the shear cracks formed at 45º. However, they got deflected 6along the reinforcement after reaching the links. There was little bridging of the cracks by the 7links. The links, in all probability, wouldn’t have yielded at that time. As a result, the length and 8density of cracks were much higher in D specimens. The crack deflection mechanism renders 9toughness to the joint. Understandably, the dissipation of energy in the D specimens would be 10higher. In the B specimens the angle of cracks was approximately 45°. Due to the absence of 11shear links in the path of the crack it split the space between the two links diagonally. This 12demonstrates the paramount importance of spacing of stirrups in the joint behavior. 13 14The initial cracking was restricted in the arm of the beam outside the joint area. At 12mm tip 15deflection the cracks started developing inside the joint area. Therefore, only the post yield 16behavior was influenced by the joint cracks. There was a major difference between the D and B 17specimens in the cracking inside the joint area. In D specimens the cracks developed diagonally 18indicating shear deformation in the joint. In B specimens, on the other hand, vertical tension 19cracks started developing both at the center of the joint and at the distant face of the column. 20This indicates the reduction in bond between the reinforcement and concrete. The presence of the 21central link in the D column made a major difference at this phase. The central link ensured a 22negligible strain along itself and therefore, it impeded tension cracks. The tension cracks on the 23junction between Arm2 and column also affected the performance of arm 2 adversely. The loss 24of bond in the B specimens resulted in rapid stiffness degradation. This is borne out by the 1 2

8

1remarkably narrower waistline of the hysteresis plot of the B specimen (Fig. 6b). The pinching of 2the hysteresis graph is a testimony of bond slip in the joint. 3 4The difference in the crack patterns in the D and the B specimens is illustrated in Fig. 9. While 5the B specimens had suffered large scale spalling of concrete in the joint region the D specimens 6did not undergo large spalling and therefore, retained their strengths even after a large 7deformation. 8Ability of the structure to survive an earthquake depends to a large extent on its capability to 9dissipate the input energy. An estimate of the hysteretic damping can be found by the area 10enclosed in the load-displacement hysteresis loops (Fig. 10). It may be noted that a wider loop 11(i.e. a large difference in ordinates in the ascending and the descending paths) would signify 12higher hysteretic damping. Cumulative energy dissipated was calculated by summing up the 13energy dissipated in consecutive loops through out the test. The energy dissipation curves for D 14and B specimens are shown in figure 10. The stepped nature of the curves is due to the repetition 15of the same displacement level three times in each load epoch. It indicates that the incremental 16damage, and therefore energy dissipation of energy, takes place in the first cycle of the epoch. 17There is no significant energy dissipation in the following cycles of the same epoch. The D joint 18exhibits much higher dissipation of energy than the B joint. A quantitative analysis of their 19performance is included later. 20 21Brittle Specimens at Different Axial Loads 22To understand the effect of column compression on joint behavior the brittle specimens are 23tested at 10, 25 and 37% of axial compression capacity of the columns. The Load deflection 24envelopes have been presented in Fig. 11. The peak load increases with increase in the axial load 1 2

9

1as shown in figure 11. However, the stiffness degradation after the peak load was much more 2sudden in the columns with higher axial compression. As a result, contrary to expectation, the 3energy dissipation did not go up with the increase in the column compression. 4To understand the anomaly of the joints with heavily loaded columns we shall compare the crack 5histories in Figs.8b and 12. The joint tension crack appeared later in B25 than in B10. The joint 6remained unaffected in case of B37. As a result, the loss of bond of the beam tensile bars 7developed later in B25 and the bond was not lost at all in B37. Hence, the prepeak stiffness and 8the peak load are higher in the heavily compressed columns. However, the bond slip also leads to 9postpeak dissipation of energy through friction. Therefore, the postpeak slope of the envelope is 10more gradual in B10 than B25 and B37. As we increase the axial load on columns we would 11expect the failure to shift from the beam to the column. 12Fig. 13 shows the joints B25 and B37 at severely damaged state. Both specimens had major 13damage due to shear outside the joint core area. The shear capacity of the beam was reached in 14this case due to higher peak loads. The diagonal shear cracks coalesced forming triangular 15wedges that spalled and exposed the reinforcements almost entirely. The shear failure was 16sudden and the samples lost strength rapidly once the triangular wedges started spalling. In one 17case a large portion of the column cover concrete had spalled (B37). The spall was due to the 18higher bursting pressure as a result of higher compressive stress in column, as well as the tension 19in the beam longitudinal reinforcements. Fig. 14 illustrates the energy curves for the specimens 20with varying axial compression. The dissipated energy reduces with the increase in the column 21compression. The spall in specimen B37 has reduced the energy dissipation severely. This 22highlights the importance of confining concrete both in beams and columns in the vicinity of the 23joint. 24 1 2

10

1Comparison of Arm 1 and Arm 2 2The joints were cast with two arms. The arms were tested sequentially. The objective of the test 3was to estimate the performance of a heavily damaged joint. Fig. 15 shows the load-deflection 4hysteresis of all the joints. It can be seen that the peak load level in case of Arm-2 has reduced 5for all the joints. The pinching of the hysteresis curves is also more evident in Arm2. This 6indicates that the bond slip is more in that arm. The difference in behavior of the two arms is 7much more in the D specimen than the B specimens. The Damaged ductile specimens behave 8quite similar to the brittle specimens. 9Table 3 presents the loads and displacements at the yield point and peak load point for all the 10specimens. Arm2 of all the specimens had lower peak yield load and peak load. As a result, the 11energy dissipation in Arm2 was much lower than in Arm1. There is a more prominent reduction 12in energy dissipation in the D specimen than the B specimens. 13

CONCLUSION

14Following major conclusions are made from the present investigation: 15

•

The beam-column joints with closely spaced shear links (ductile) have superior postyield

16

behavior than the joints sparse links (brittle) due to better confinement of concrete and

17

crack deflection mechanism. The longer and denser shear cracks in the ductile joints lead

18

to higher energy dissipation.

19

•

The links in the core area of ductile joints prevents bond slip and therefore, increases the

20

hysteretic damping. The pattern of cracking in the core area of the joints is different for

21

the ductile and the brittle joints.

22

•

The peak load goes up with the increase in the column compression due to the

23

prevention of bond slip by the higher compressive force. However, the failure is much

1 2

11

1

more sudden in joints with high column compression. The energy dissipation of joints

2

with high column compression is markedly lower.

3

•

The loss of capacity in strength and energy dissipation in the damaged ductile specimens

4

is much higher than that in the damaged brittle specimens. This is due to more

5

pronounced bond slip in the damaged ductile specimens than the fresh ones.

6 7

ACKNOWLEDGMENT

8

The present work is financially supported by the Board of Research in Nuclear Sciences. Dr.

9

G. Rami Reddy has helped with the instrumentation for the experiments. The experiments are

10

carried out at the Structural Integrity Testing and Analysis Centre of Indian Institute of

11

Technology Bombay, Mumbai, India. The authors would also like to thank M/s Fyfe India

12

for supplying the composite material system.

13 14

REFERENCE

15 161. Hanson NW, Connor HW. Seismic Resistance of Reinforced Concrete Beam-Column Joints. 17

J Structural Division Proceeding of the American Society of Civil Engineers 1967;93:533-

18

560.

192. Hanson NW. Seismic Resistance of Concrete Frames with Grade 60 Reinforcement. J 20

Structural Division Proceeding of the American Society of Civil Engineers 1971;97:1685-

21

1700.

223. Lee DL., Wight JK. Hanson RD. RC Beam-Column Joints under Large Load Reversals, J 23

1 2

Structural Division, proceedings of the ASCE 1977;103:2337-2350.

12

14. Filippou FC, Popov EP, Bertero VV. Analytical Studies of Hysteretic Behavior of R/C 2

Joints. J Structural Division ASCE 1986;112:1605-1622.

35. Durrani AJ, Wight JK. Behavior of Interior Beam-to-column connections under Earthquake4

Type Loading. ACI J 1985;82:343-349.

56. Meinheit DF, Jirsa JO. Shear Strength of R/C Beam-Column Connections. J Structural 6

Division AMCE 1981;107:2227-2244.

77. Clyde C, Pantelides CP, Reaveley LD. Performance-Based Evaluation of Exterior Reinforced 8

Concrete Building Joints for Seismic Excitation. Pacific Earthquake Engineering Research

9

Report :Berkeley: University of California; 2000/05, 2000.

108. Abrams DP. Scale Relations for Reinforced Concrete Beam-Column joints. ACI Structural J 11

1988;84:502-512.

129. Park R, Ruitong D. A Comparison of the Behavior of Reinforced Concrete Beam- Column 13

Joints Designed for Ductility and Limited Ductility. Bulletin of the New Zealand National

14

Society of Earthquake Engineering, 1998;21(4),1998:255-278

1510. Lowes LN, Moehle JP. Evaluation and Retrofit of Beam-column T-Joints in older Reinforced 16

Concrete Bridge Structure. ACI Structural J 1998;96:519-532.

1711. Dhakal RP, Pan TC, Irawan P, Tsai KC, Lin KC; Chen CH.. Experimental Study on 18

Dynamic Response of Gravity Designed RC Connections. Engineering Structures 2005;

19

27(1):75 – 87.

2012. Mukherjee A, Joshi M. FRPC Reinforced Concrete Beam-Column Joints under Cyclic 21

Excitation. Composite Structures 2005;70:185-199.

2213. IS: 13920. Code of Practice for Ductile Detailing of Reinforced Concrete Structures. New 23

1 2

Delhi: Bureau of Indian Standard; 1993.

13

114. IS: 456. Code of Practice for Plain and Reinforced Concrete. New Delhi: Bureau of Indian 2

1 2

Standards; 2000.

14

1 2

TABLES AND FIGURES

3List of Tables: 4Table 1— Properties of Reinforcing Materials 5Table 2— Test Matrix 6Table 3— Percentage increase and comparison in arms 7 8List of Figures: 9 10Fig. 1— Failure of RC Joints in Kutch Earthquake 11Fig. 2— Stress- Strain Behavior of Concrete 12Fig. 3— Details of test specimens 13Fig. 4— Experimental setup 14Fig. 5— Displacement Cycle 15Fig. 6— Load-displacement curves for D and B-Type specimens 16 17Fig. 7— Envelope Curves for D-Type and B-Type Specimens 18Fig. 8— Damage history of D and B –Type specimens 19Fig. 9— Cracks in ductile and brittle specimens 20Fig. 10— Energy curves (D-10 and B-10) 21Fig. 11— Combine Envelope Curve (B-Type arm-1) 22Fig. 12— Damage history (B-25 and B-37) 23Fig. 13— Damage in specimens with higher column compression 24Fig. 14— Energy curves (B-10, B-25 and B-37 specimens) 25 26Fig. 15: Load versus displacement curves for D and B specimens 27 1 2

15

1 2

Table 1— Properties of Reinforcing Materials

3 4

Material

Diameter (mm)

Steel bar longitudinal reinforcement Steel bar transverse Reinforcement

5 6 7 8

Tensile Modulus (GPa) 154

Ultimate Strain

16 and 12

Tensile Strength (GPa) 0.65

8 mm

0.5

193

0.043

0.062

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 1 2

Table 2— Test Matrix Specimen

Axial force to column (% of column capacity)

D-10 D-10-2 B-10

10 % 10 % 10 %

B-25 B-37

25 % 37 %

16

1 2

Table 3— Percentage increase and comparison in arms

3 Specimens

Arms

Arm 1 D-10 Arm 2 Arm 1 B-10 Arm 2 Arm 1 B-25 Arm 2

B-37

Arm 1 Arm 2

Direction Up

Yield point Load Deflection (kN) (mm) 32.01 5.339

Down

-28.68

-5.32

Up

23.80

9.1

32.74

25

Down

-24.32

-9.6

-33

-24

Up

27.05

5.97

30.94

10.89

Down

-22.97

-6.35

-31.94

-16.02

Up

20.71

6.82

25.12

12.61

Down

-19.17

-7.05

-24.45

-12.82

Up

30.23

9.01

36.88

19.78

Down

-32.05

-7.58

-36.66

-13.45

Up

28.16

6.19

34.48

12.19

Down

-28.07

5.93

-36.12

-13.93

Up

34.56

5.4

38.11

11.45

-30

-5.76

-35.14

-8.56

Up

31.94

11.09

36.8

21.85

Down

-33.97

10.78

-35.88

22.3

Down

Peak load point Energy Load Deflection Dissipation (kN-m) (kN) (mm) 35.05 17.49 23.89 -34.02 -16.74 9.22 8.45 6.19 8.16 7.41 4.87 4.21

4 5 6 7

1 2

17

1

(a) Confinement failure

(b) Shear failure

(c) Combined lap and shear failure

Fig.1— Failure of RC Joints in Kutch Earthquake 2 3 4 5

24

Stress N/mm

2

20 16 12 8 4 0 0.0000

0.0004

0.0008

0.0012

Strain

6 7 8

Fig. 2— Stress-Strain Behavior of Concrete

9 10 11 12 1 2

18

1 2

800

800

650

.. ..

150

.. ..

150

650 200

50

Stirrup 8 Dia 75 mm c/c

200

12 Dia bar Stirrup 8 dia 150 mm c/c

12 Dia bar Stirrup 8 Dia 150 mm c/c

16 Dia bar 200

16 Dia bar

200

3 4

(a) D-Type

(b) B-Type

5Note: All dimensions are in mm Fig. 3— Details of test specimens 6. 7 8

Load cell (Column axial load) Actuator LVDT 5

Load cell (Cyclic loading) LVDT 2

LVDT Hydraulic Jack

100

200

Actuator

160

LVDT 3 430 LVDT 4

LVDT 1 640

9 10 11 12 13

1 2

(a) Test setup

(b) Instrumentation Fig. 4— Experimental setup

19

60

Displacement (mm)

40 20 0 -20 -40 -60 0

1000

2000

3000

4000

5000

6000

Time (s)

1 2

Fig.5— Displacement Cycle

3 4

40

40

30

30

5

20 load (kN)

20

6 7 8

load (kN)

10 0

10 0

-10

-10

-20

-20

-30

-30

-40 -40

-30

-20

-10

0

10

20

Displacement (mm)

30

40

50

-40 -40

-30

-20

-10

0

10

20

30

40

Displacement (mm)

9 10

(a) D-10 (arm-1)

(b) B-10 (Arm-1)

11 12 13 14 15 16 17 18

1 2

Fig. 6— Load-displacement curves for D and B-Type specimens

20

40

D-10

30 20

B-10

Load KN

10 0 -10 -20 -30 -40 -40

-30

-20

-10

0

10

20

30

40

Displacement (mm)

1 2 3 4 5 6 78

Fig. 7— Envelope Curves for D-Type and B-Type Specimens

4 mm

20 mm

8 mm

6 mm

10 mm 8 mm 14 mm

14 mm

9 10

12 mm

2 mm

(a) D-10 (arm-1)

12

2 6 mm

4 m m

(b) B-10 (arm-1)

Fig. 8— Damage history of D and B –Type specimens

11 12

1 2

21

1 2 3

(a) D-10 (arm-1)

(b) B-10 (arm-1) Fig. 9— Cracks in ductile and brittle specimens

4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 1 2

22

1 2 3 4

6

20

Energy (kN-M)

5

D-10-Arm-1

25

15

10

B-10-arm-1

7 8 9

5

0 0

5

10

15

20

25

30

35

40

Displacment (mm)

10

Fig. 10— Energy curves (D-10 and B-10)

11 12 13 14 15

B-25

40

B-37

30

16

18

Load (KN)

17

20

B-10

10 0 -10 -20

19 20

-30 -40 -30

-20

-10

0

10

20

30

Displacement (mm)

21 22

Fig. 11— Combine Envelope Curve (B-Type arm-1)

23 24 1 2

23

1

16 mm 10 mm

6 m m

8 mm 14

4 mm

2

(a) B-25(arm-1)

3

2 mm

12 mm mmm m

4 2 mm m m 134

14 mm

12 mm mmm m

(b) B-37(arm-1) Fig. 12— Damage history (B-25 and B-37)

4

5 6 7 8

1 2

(a) B-25

(b) B-37

Fig. 13— Damage in specimens with higher column compression

24

1 2 3 B-25

10

B-10

4 5 6 7

Energy (kN-M)

8

6

B-37

4

2

8 0

9 10 11 12 13

1 2

0

4

8

12

16

20

24

28

Displacement (mm)

Fig. 14— Energy curves (B-10, B-25 and B-37 specimens)

25

1 40

27 28 29 30 31 32 33 34

Arm-1

40

20

10

10

0

Load (KN)

load (kN)

Arm-2

30

20

-10 -20

0 -10 -20

-30 -40 -40

-30

-20

-10

0

10

20

30

40

50

-30

(a) D-10

-40 -30

-20

Displacement (mm) 30

-10

0

10

20

30

Displacement (mm)

30

40

20

20 10

Load (KN)

load (kN)

10

0 -10

(b) B-10

-20 -30

0

-10

-20

-30

-40 -40

-40

-30

-20

-10

0

10

20

30

40

40

30

30

20

20

10

Load (KN)

Load (KN)

-30

-20

40

Displacement (mm)

10 0 -10

-10

0

10

20

30

40

Displacemnt (mm)

0 -10 -20

(c)B-25

-20

-30

-30

-40

-40 -30

-20

-10

0

10

20

-30

30

-20

40

40

30

30

0

10

20

30

20

(d) B-37

10 0

Load (KN)

20

10 0

-10

-10

-20

-20

-30

-30 -40

-40 -30

-10

Displacment (mm)

Displacement (mm)

Load (KN)

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

30

-20

-10

0

10

Displacement (mm)

20

30

-20

-10

0

10

20

30

40

50

Displacement (mm)

Fig. 15: Load verses displacement curves for D and B specimens

35 36 37 38 39 1 2

26